Findings: Expert Consensus on a Proposed Framework to Explore Factors Influencing Malaria Preventive Behavior Using a Modified Delphi Method

This study employed a modified Delphi method to establish expert consensus on a study framework aimed at identifying factors influencing malaria preventive behavior in communities exposed to Plasmodium knowlesi malaria. The Delphi method, recognized for its capacity to gather expert judgments on complex and underexplored topics, was selected due to its flexibility and iterative feedback mechanism. The study sought to achieve consensus, defined as a 70% agreement threshold, among participants on items integral to the proposed framework.

Introduction

The Delphi method, originally developed in the 1960s by the Rand Corporation, has been widely applied across various disciplines, including healthcare, to inform decision-making, evaluate concepts, and develop study frameworks. It combines both qualitative and quantitative elements, relying on structured communication among a panel of experts to synthesize opinions and achieve consensus. Its reliability stems from the iterative process of controlled feedback, where participants revise their responses based on anonymized summaries of group opinions.

In healthcare research, the Delphi method is particularly suited for scenarios where experimental methods are impractical or when existing evidence is insufficient. Despite some critiques regarding its lack of universal methodological standards and potential for biases, it remains a trusted approach for generating frameworks, guidelines, and actionable insights. This study adapted the Delphi method to address factors influencing malaria preventive behaviors, integrating theoretical models and systematic evidence to shape its structured questionnaire.

Methods and Analysis

The study aimed to establish a consensus on factors that should be included in a framework for exploring malaria preventive behavior in affected communities. A pragmatic approach guided the methodology, with participants selected based on expertise, publication history, and professional experience in malaria research. Purposive and snowball sampling ensured a diverse group of academics and practitioners, while stringent inclusion criteria guaranteed the credibility of expert contributions.

A structured, multi-round process facilitated the consensus-building exercise. The study was conducted electronically, leveraging ICT tools like email and Google Forms to accommodate the geographical dispersion of participants. This approach minimized logistical constraints, enhanced confidentiality, and reduced participant fatigue. The iterative rounds were designed to span five months, with each round lasting approximately four weeks. Participants were reminded weekly to submit responses, ensuring timely progress.

Study Procedure

The Delphi process began with the development of a structured questionnaire informed by an extensive literature review. Items for inclusion were derived from behavioral theories, including the Ideation Model of Communication and Behavior Change, Murdock’s Model of Illness Causation, and Kleinman’s Explanatory Model. These theoretical frameworks provided a robust basis for understanding cognitive, emotional, and social factors influencing behavior.

The survey underwent rigorous validation through content and face validation, followed by a pilot study involving seven volunteers with expertise in malaria research. Feedback from the pilot study was used to refine the questionnaire, ensuring clarity, relevance, and methodological rigor.

Round 1 introduced the validated survey to participants, collecting both quantitative responses on a 5-point Likert scale and qualitative feedback to capture additional insights. Demographic data, including participants’ gender, job position, and years of experience, were also recorded. Items achieving a 70% agreement rate and a median score of 4 or higher were advanced to subsequent rounds, while those failing to meet this threshold were revised or discarded.

In Round 2, participants reviewed summarized responses from the first round. The summaries included aggregated ratings and thematic analyses of qualitative data. Participants were encouraged to reassess and, if necessary, revise their initial responses based on the group’s collective feedback. This round also excluded open-ended responses, focusing instead on refining and prioritizing items that reached consensus.

Round 3 served as the final stage, presenting participants with consolidated results from the earlier rounds. Items retained in this stage represented those that met the predefined 70% consensus threshold. Participants received individualized and group feedback, allowing for a final opportunity to confirm their agreement or propose adjustments. The iterative design ensured that only well-supported items were included in the final framework.

Results and Framework Development

The Delphi process successfully achieved consensus on key factors to be included in the study framework. These factors were categorized based on theoretical models, reflecting cognitive, emotional, and social influences on malaria preventive behavior. The framework also incorporated local beliefs about illness, highlighting the significance of cultural and contextual determinants.

Statistical analysis using SPSS and thematic analysis for qualitative data ensured a comprehensive evaluation of responses. The use of medians to assess central tendencies allowed for a clear representation of consensus, with higher medians indicating stronger agreement among participants.

Challenges and Considerations

Developing the survey posed several challenges, including defining questions that balanced depth and brevity. Ensuring participant engagement across iterative rounds required careful planning, as false consensus and fatigue are common risks in Delphi studies. The study addressed these issues by limiting the number of rounds to three and maintaining transparency about the process and objectives.

The selection of experts and the sample size were also critical to the study’s success. While a minimum of 12 participants was deemed sufficient to capture diverse perspectives, the study emphasized the importance of ensuring representation across different professional disciplines and geographic contexts.

Conclusion

This modified Delphi method effectively synthesized expert opinions to develop a study framework for exploring factors influencing malaria preventive behavior in communities exposed to Plasmodium knowlesi. By integrating theoretical models, rigorous validation processes, and iterative rounds of feedback, the study ensured the reliability and relevance of its findings. The resulting framework provides a foundation for future research and interventions aimed at improving malaria prevention efforts.

The study highlights the utility of the Delphi method in addressing complex public health challenges, offering a structured and inclusive approach to achieving consensus on underexplored issues. While the method requires careful design and management to address its inherent limitations, its ability to generate actionable insights from diverse expertise makes it a valuable tool in health sciences research.